This is a longitudinal study proposing to follow elderly individuals over time to determine the morbidity and mortality associated with sleep-related respiratory disturbance (SRRD). SRRD shows a strong age-related prevalence and indirect evidence suggests that SRRD could be associated with higher mortality. Likewise some studies have suggested associations between SRRD and morbidity (impairment in psychological and cardiovascular function), but the findings to date have been inconclusive. For cardiovascular function in particular, studies have not employed epidemiological definitions of cardiovascular disease. In this study, two samples of aged individuals will be followed over time: 1) a group of 150 elderly former sleep clinic patients with documented SRRD at entry and 2) a group of 100 non-clinic elderly research subjects studied biennially. Respiration in sleep will be recorded for 2 nights with in-lab polysomnography in sample 1 and with ambulatory monitors for 2 nights in sample 2. All individuals will also undergo a psychological evaluation (consisting of measures of depression, psychological symptoms, neurological impairment, and a newly-validated inventory of self-reported daytime sleep tendency) and a cardiovascular evaluation (consisting of 12-lead electrocardiography, and standardized measurements of myocardial enzymes, angina and hypertension). Primary analyses will focus on prediction of putative outcomes through use survival analysis in a proportional hazards model. Follow-up interval and other measures at entry (SRRD, age, sex, obesity) will serve as risk factors. By using previously acquired data, the proposed study is a cost- effective means to investigate the progression of SRRD over time. If associations between SRRD and morbidity outcomes can be demonstrated here, full-scale prospective epidemiological studies of SRRD in groups more representative of the general population would be implicated.